| Message ID | 20240229154941.99557-6-xni@redhat.com (mailing list archive) |
|---|---|
| State | Superseded, archived |
| Delegated to: | Mike Snitzer |
| Headers | show |
| Series | Fix dmraid regression bugs | expand |
diff --git a/drivers/md/md.c b/drivers/md/md.c index f264749be28b..cf15ccf0e27b 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -6341,6 +6341,7 @@ static void __md_stop_writes(struct mddev *mddev) void md_stop_writes(struct mddev *mddev) { mddev_lock_nointr(mddev); + set_bit(MD_RECOVERY_FROZEN, &mddev->recovery); __md_stop_writes(mddev); mddev_unlock(mddev); }
After patch commit f52f5c71f3d4b ("md: fix stopping sync thread"), dmraid stops sync thread asynchronously. The calling process is: dev_remove->dm_destroy->__dm_destroy->raid_postsuspend->raid_dtr raid_postsuspend does two jobs. First, it stops sync thread. Then it suspend array. Now it can stop sync thread successfully. But it doesn't set MD_RECOVERY_FROZEN. It's introduced by patch f52f5c71f3d4b. So after raid_postsuspend, the sync thread starts again. raid_dtr can't stop the sync thread because the array is already suspended. This can be reproduced easily by those commands: while [ 1 ]; do vgcreate test_vg /dev/loop0 /dev/loop1 lvcreate --type raid1 -L 400M -m 1 -n test_lv test_vg lvchange -an test_vg vgremove test_vg -ff done Fixes: f52f5c71f3d4 ("md: fix stopping sync thread") Signed-off-by: Xiao Ni <xni@redhat.com> --- drivers/md/md.c | 1 + 1 file changed, 1 insertion(+)